The
Sydney Morning Herald reports that the former NSW Treasury secretary Michael Lambert has been conducting an audit of the state's finances for the recently elected (one year ago) Liberal/National coalition (centre right) administration (which has followed a 16 year long Labor administration). His work was finished last year, but received little attention, apparently because the state government has lost interest in reform.
One of his key findings was that "vehicle taxes" be replaced with a Sydney wide congestion targeted urban road pricing scheme, which would apparently generate economic benefits of A$720 million (US$747 million) per annum through reduced delays. The news report says nothing more, but the complete report on public finances in New South Wales is available on the
NSW Treasury website here. There is much excellent analysis here for anyone interested in public finances.
What does the report say?
The relevant sections are in Chapter 13, and is worthwhile reading for those who still think that vehicle ownership taxes and fuel taxes are optimal ways of charging for road use.
The report analysed the "excess burden" of taxes, being the cost of taxes to the economy, over and above the revenue acquired by the government. These costs are deadweight losses in reduced economic activity because of the loss of consumer welfare from the higher price of what is taxed.
Taking those costs, administrative costs and the wider economic costs and benefits of different ways of charging road vehicles produces some fascinating conclusions.
Vehicle registration tax creates an excess burden of A$474 million (US$492 million) per annum for the state, which is a loss of 25% of the value of the revenue collected from it. Remember that, because that loss is what can be theoretically abolished with economically efficient road pricing.
The report proposes replacing taxes on vehicle ownership with a state-wide flat rate distance road user charge, with charges in Sydney varying by time and location so that congestion could be targeted. The light vehicle charge would be A$0.04 per km (US$0.07 per mile). Sydney roads would, at certain times (i.e. peaks and interpeaks), have a congestion surcharge of between A$1.15 and A$3.83 per km (US$1.91 and US$6.39 per mile).
It calculated that the net economic welfare gain for the state would be A$662 million per annum (US$687 million). Another A$5 billion per annum (US$5.2 billion) is saved from reduced congestion and road maintenance costs.
A$2.5 billion (US$2.6 billion) is the revenue needed to replace ownership/transfer taxes.
A$780 million (US$810 million) of revenue alone could be generate by tolling the existing untolled portions of Sydney's orbital motorway network, using existing average prices, with peak and offpeak rates.
A national (beyond New South Wales) heavy vehicle distance based road user charge would raise an additional A$100 million (US$104 million) a year for NSW from out of state trucks using NSW roads, if it replaced heavy vehicle registration fees. That's because a significant amount of km are driven on NSW roads by non-NSW registered trucks.
A NSW only heavy vehicle weight and distance road user charge priced to fully meet the marginal costs of heavy vehicle road use, would raise A$1 billion per annum in revenue (this would also replace fuel tax for heavy vehicles).
The remainder would come by charging light vehicles across the rest of the state, including the non-motorway network in Sydney.
While administrative costs are estimated to cost anywhere between 10 and 25%, so charges would have to recover A$2.5 billion in revenue AND an additional A$250-A$750 million.
This is a common argument against road pricing, except the report points out that introducing road pricing more than offsets the administrative costs by:
- Eliminating the deadweight excess burden of existing taxes would save 26% of the cost of that revenue now from the economy. That net gain does not accrue to the Treasury of the state government, but is distributed throughout the economy.
- Significantly reducing the costs of congestion, producing gains of hundreds of millions per annum in time savings.
Congestion pricing, if applied on a network wide basis, charging by time, place and distance, set at the rates above to efficiency target congestion would also generate another A$2 billion (US$2.1 billion) in additional net revenue.
Instead of network pricing, cordon pricing (applied at A$10 to enter any of Sydney's 13 business districts) could generate less revenue, but risks creating congestion outside those districts and is considered to not be suitable for Sydney.
Expanding tolling on major routes is also seen as unsuitable because it would divert traffic onto other routes.
The conclusion from the report is that there are considerable merits in shifting away from current forms of motoring taxation to road pricing.
The first step proposed is to expand tolling on Sydney motorways at prices that reflect current averages on other tolled motorways. The report proposes it be followed by renegotiations of existing concessions so that toll pricing can target peaks and have off peak discounts. It also proposes serious consideration be given to a NSW only heavy vehicle distance charging system, which if possible is consistent with a future Australia wide system. This should be followed by piloting of distance charging as an option to start a transition from existing taxes.
My view
This report is one of the more economically sound pieces of analysis on road pricing as complete replacement source of revenue from road users I have seen in some time. It is more telling that the report actually came to this conclusion not as a road pricing study, but as a wider economic study into the entire taxation, expenditure and state treasury administrative, procedural and governance framework.
It moves beyond the purely financial analysis of taxation, that almost always demonstrates that existing taxes have low external costs, by raising the deadweight cost of taxation and the external benefits that other forms of charging can generate by addressing inefficiencies such as congestion.
The point is this. Existing motoring taxes create distortions of their own, because they do not reflect the infrastructure and economic costs of providing roads. They tax ownership and sale of vehicles (and although not included in the analysis, fuel tax does not charge road use).
Road pricing has an administrative cost higher than existing revenue sources, but its economic cost is lower and on top of that it can create net economic benefits by reducing congestion and road damage, because of changes in behaviour.
The steps proposed to move forward are worthy of further consideration. Heavy vehicle charges would be a logical and manageable step forward that would generate most of the road damage reduction benefits, a few congestion reduction benefits, but also help de-risk a wider transition. Expanding tolls in Sydney would appear easy, but would need to be done in a way to minimise diversion risk. The bigger leap towards shifting cars to distance charging should be done through a voluntary system to start with.
The one yawning gap in this report is fuel tax, which in unsurprising as it is levied at the Federal level, so is a Federal problem (states get a share through grants from the Federal government rather than directly receiving tax revenue). Fuel tax is likely to create its own deadweight burdens, although it arguably has modest positive impacts on environmental outcomes.
It would be a shame if politics means that nothing is done about the road charging portion of this report in New South Wales. However, whatever happens there I hope the deadweight burden of taxation is weighed into assessment of motoring taxes elsewhere. What this report does is to help place Australia in the ranks of countries which are seriously considering the full economic impacts of existing forms of taxation of road use. Let the debate continue...
Other interesting facts from the report
A$138 million (US$143 million) was raised from tolls on state roads (largely Sydney Harbour Bridge). This revenue has been growing at a rate of 9.7% per annum (nominal) over 10 years.
A$87 million (US$90 million) was raised from "plate fees" on state roads, meaning charges for those not paying tolls with accounts, but with surcharges after their number plates were detected on free flowing toll roads. This has been growing at 12.4% per annum (nominal) over 10 years.
A$1.7428 billion (US$1.79 billion) was raised from motor vehicle registration and ownership taxes in the state in the year 2010-11. Note the state has a population of 7.3 million with 5.6 million registered vehicles.
A$590 million (US$612 million) was raised from motor vehicle transfer duty (tax on selling vehicles, separate from retail taxes).
A$95 million (US$99 million) was raised from the
parking space levy which is a tax on off-street commercial parking in specific parts of Sydney. Effectively a workplace parking levy of the kind sometimes proposed as an alternative to congestion pricing.
A list of highway PFPs (Privately Financed Projects) in the state is presented in Table 1.4.4 (7 toll roads) for those interested.